[PyTorch Edge] Allow >2 dimensional tensors in internal Quantized Matmul (#73246)
Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/73246
Iterate over the outer dimensions. Will be optimized with parallelization in future diff(s)
ghstack-source-id: 151250869
Test Plan:
From fbcode:
```buck test caffe2/test:quantization -- test_qmatmul```
Performance Improvement/Benchmarking: See testplan of D34012771
Reviewed By: kimishpatel
Differential Revision: D33735479
fbshipit-source-id: b7908e1e78a4ea0c4910fc540c5549e4ec3b7473
(cherry picked from commit 36c57cdfe81bfd9139111003f363112f123f4e6e)
